The Only Rehearsal

The success threshold was 73 seconds.

Before NASA's DART spacecraft hit the asteroid moonlet Dimorphos on 26 September 2022, the mission had a number that would count as a win: the moonlet's orbit around its parent asteroid, Didymos, had to get shorter by at least 73 seconds. On 11 October, NASA and the Johns Hopkins Applied Physics Laboratory reported that it had got shorter by 32 minutes, give or take two. The orbit had been 11 hours and 55 minutes. Now it was 11 hours and 23. The mission had beaten its own bar more than 25 times over.

ESA now rounds the change to about 33 minutes. Whichever you use, ESA's mission manager describes Dimorphos as the first object in the Solar System to have its orbit measurably shifted by human action. This month a second spacecraft arrives to find out how.

This is the second "Reality, reviewed": real technology scored the way we score fiction, as a work and as a prophecy.

The work

Strip away the engineering and DART is a very simple film. A spacecraft launched on a Falcon 9 on 24 November 2021 flies for ten months at a small moonlet and hits it at about 6 km/s. ESA's figure is 6.1. The moonlet, Dimorphos, was described by NASA at the time as 160 metres across; ESA now gives 151. It was chosen because it orbits a bigger asteroid, 780 metres across, which makes its motion easy to time from Earth with ordinary telescopes. Nobody had to guess the result. Astronomers just had to watch the eclipses.

The whole project cost about $325 million, by the Planetary Society's tally of the budget. For that we got a measurement, and measurements are what this field had been short of. The Italian Space Agency's cubesat LICIACube, released from DART before the impact, flew past Dimorphos and watched it.

The surprise is the number. The impact transferred more momentum than the spacecraft carried, because debris thrown off the surface pushed back like exhaust. The ratio, called beta, came out at 3.61 in the main Nature paper (plus 0.19, minus 0.25), assuming Dimorphos and Didymos have the same density. In other words, impact and ejecta together shoved the moonlet about three and a half times harder than the spacecraft's own momentum would have. NASA's own Lori Glaze called the 32 minutes "consistent with our estimates beforehand, but clearly at the upper end of that range". ESA, writing this autumn, is blunter: it asks which properties of the asteroid explain "a deflection so much greater than impact simulations predicted".

That is the caveat that keeps the score from going higher. A beta of 3.6 assumes a density we don't know. The same paper allows anything from 2.2 to 4.9 depending on the mass of Dimorphos, because nobody has weighed it. One impact on one pile of rubble is a proof of concept. It isn't a method.

And it is a very good proof of concept, with a team that knew exactly what it was doing and said so honestly. On the art, I'd give it nine. I've knocked it down to 8.3 because a test isn't a defence.

The sequel

The reason this is on the front page this month is that the sequel is arriving. ESA's Hera launched on 7 October 2024. Its job is the crash-site investigation: how big is the crater, did the asteroid reshape itself, what is it made of, how much does it weigh. On 18 October it took the picture that matters for the story, from 1,280 km out, in which Dimorphos was about one pixel, but a pixel clearly apart from Didymos. The probe had slowed itself to 20 m/s relative to the pair with two thruster firings a week apart. ESA's mission manager, Ian Carnelli, says the first detailed images of Dimorphos will be shown on 25 November.

Then two shoebox-sized cubesats are due to go closer, and the plan ends, eventually, with a landing. If Hera pins down the mass, the biggest uncertainty in beta shrinks. That's the difference between "it worked" and "we could do it again to something that was aimed at us".

The prophecy: what the film told us to worry about

It's worth setting DART against Don't Look Up, because the film got it half right. Adam McKay's 2021 comedy, released a few weeks after DART launched, gives its scientists about six months before a comet hits. It was meant as an allegory for climate change, and it is more about the media and the politicians than about the comet. In it, Rob Morgan plays the head of the Planetary Defense Coordination Office, which is a real NASA office, and the one that funded DART.

The film's fear was that we'd know and not act. The real record so far is the other way round. When the act was a small, well-defined engineering job, it got done, on schedule and under a threshold nobody expected to be beaten. The weak points are earlier and later.

Earlier: finding the rock. You can only deflect what you've found, and the survey of the big ones is unfinished. Congress mandated a search for 90 per cent of near-Earth asteroids larger than 140 metres in 2005. NASA's dedicated infrared telescope for it, NEO Surveyor, is now scheduled to launch no earlier than September 2027, after a succession of proposals that weren't picked for launch between 2006 and 2017. The small ones arrive unannounced. The object that burst over Chelyabinsk in February 2013 was about 17–20 metres across, and nobody saw it before it hit the air.

2024 YR4 was the closest this came to a live exercise. It was found on 27 December 2024. By 18 February 2025 its chance of hitting Earth in December 2032 peaked at 3.1 per cent; by 23 February new observations had taken its Torino rating to zero. For a while it also had a small chance of hitting the Moon, which NASA ruled out in March 2026 with Webb observations. All of that happened without anyone launching anything. That's the good news. I couldn't find any deflection mission begun for it, and had the odds gone the other way, I could find no interceptor ready to fly. DART is a proof; it is not a defence on a pad.

Later: who decides. The international machinery exists, and it's more serious than most people know. The UN-endorsed International Asteroid Warning Network handles detection and warning. The Space Mission Planning Advisory Group, set up by space agencies on a UN mandate, develops reference plans for a deflection campaign. Both coordinate and recommend, as the UN and ESA describe them. Nothing I could find gives either of them the power, or the money, to launch. A deflection is, in the end, a national decision about a global hazard, and the first one will be a political event before it is an engineering one. That is the part Don't Look Up was really about, and it is the part DART can't test.

Nearness: ARRIVING. The push is real. 2029 will be the next rehearsal: on 13 April 2029 the asteroid Apophis passes about 31,600 km above Earth, close enough that it should be visible to the naked eye in parts of Europe, Africa and western Asia, and both NASA's OSIRIS-APEX and ESA's Ramses are planned to study it.

What to do with it

Go and look at 25 November, when Hera's first sharp images of Dimorphos are due. It will be the first time we see what a thing we did to the sky looked like up close.

And watch the budget line for the survey telescope, which is the part nobody films. We have shown we can move a rock. The next test is whether we get around to looking for the ones that matter before we need to.

We rehearsed the end once, and it went well. It's the only rehearsal we've had.